Program & Abstracts - McGill PhysicsSPM/nano2012/cifar_rqmp_nano2012_progra… · Program &...

&

May , 2012 at McGill University in 21-25 MontrealNanoelectronics Summer School 2012

Program & Abstracts

A

D

F

B B

B C

E

E

A A

A A

AMcConnell Engineering Bldg3480 rue University, room 13

Rutherford Physics Building3600 rue University, board room

Trottier Building3630 rue University, lobby

New Residence Hall3625 avenue du Parc, lobby

Centre Mont-Royal2200 rue Mansfield

Restaurant L'Academie2100 rue Crescent

B

C

D

E

F

INVITED TALKS

Kirk Bevan Department of Mining and Materials Engineering, Department of Physics, McGill University

Using Effective Mass Models in Nanoelectronics

Tuesday 9:00-10:00 am

In this talk the fundamentals, applications, advantages, and limitations of the effective mass

approximation in nanoelectronic device modeling will be covered. Starting from the fundamental

finite difference approach, we will examine the importance band bending, doping, and self-

consistency in capturing device and materials phenomena. An emphasis will be placed on length

scales on the order of 100nm.

Using Atomistic Models in Nanoelectronics

Tuesday 10:30-11:30 am

In this talk the various types of atomistic models will be covered, with an overview of their

benefits and limitations. In particular, we will cover the recent rise in popularity of atomistic

models to capture bulk properties, as well as interface and surface interactions between distinct

materials in nanoelectronic architectures. The state-of-the-art will be overviewed in capturing

various materials and device structures. This will include an overview of magnetoresistance,

semiconductor-oxide interfaces, Schottky junctions, and associated phenomena. An emphasis

will be placed on length scales on the order of 10nm and below.

Some Engineering Challenges in Nanoelectronics

Tuesday 1:00-2:00 pm

In this talk we will cover some of the rising and existing engineering challenges in

nanoelectronic device design and semiconductor scaling. This will include coverage of issues

such as power consumption, leakage currents, interconnect capacitance, and next generation

memory storage. An overall emphasis will be placed on the concepts of reliability and

manufacturing variability in the maturation of any new prototype nanoelectronic technology.

INVITED TALKS

Iain R. McNab School of Biological Sciences and Applied Chemistry,

Seneca College of Applied Arts & Technology

Tuesday 2:30-4:30 pm

The Polanyi laboratory has recently been studying adsorption and reactions of molecules at

surfaces, leading to chemically attached (robust) nano-patterns.

These studies use Scanning Tunneling Microscopy to track adsorption and reaction,

augmented with first-principles calculations. The calculations are needed both to interpret the

STM images and to provide energetics of physisorption and reactive processes at the surface.

I will discuss:

1. the charge distribution at surfaces, with particular reference to the Si(111)-7x7 surface

2. modification of the surface charge distribution by physisorption and chemisorption

3. physisorption patterns of mobile and immobile species, captured by STM, and interpreted

with theory

4. mobile precursors to reaction.

Thomas Szkopek Department of Electrical and Computer Engineering, Department of Physics, McGill University

Wednesday 9:00-10:00 am & 10:30-11:30 am

Graphene - the first and most well known of the new material family of 2D atomic crystals - is

unique in its combination of material properties. In this talk, I will describe the synthesis and

electrical properties of graphene, with an emphasis on new opportunities for graphene

applications. The advent of chemical vapour deposition growth of graphene on Cu enables the

synthesis of continuous monolayer graphene films extending over macroscopic areas. Raman

spectroscopy and electron transport indicate high material quality.

The high-mobility conduction of electrons at a surface in graphene translates into the

facile measurement of surface potential. Examples of charge detection include the real-time

monitoring of surface redox reactions with atto-ampere current sensitivity, and ion trapping / de-

trapping at Si/SiO2 interfaces. Non-covalent functionalization can be used to enhance the

adsorbate sensitivity of graphene field effect transistors, including small gas molecules.

The high mobility of graphene is attractive for high-frequency applications, including

high-speed transistors, frequency multipliers and non-reciprocal devices such as Faraday

rotators. Microwave frequency transport measurements to 110GHz of graphene reveal the

complete absence of skin-effects and anomalous magneto-resistance.

My talk concludes with a discussion of the current state of the art in 2D atomic crystal

synthesis, and the potential for band engineering of 2D atomic crystals.

INVITED TALKS

Yong P. Chen Department of Physics, Purdue University

Graphene

Wednesday 1:00-2:00 pm & 2:30-3:30 pm

With many superior properties and promising application potentials, graphene has become one of

the most actively studied nanomaterials and currently attracts intense interests in condensed

matter physics, nanoelectronics, and other scientific and engineering disciplines. In these two

lectures, I will give a brief tutorial of the major material and physical properties of graphene, and

the key experimental techniques to study them, with several examples drawn from work in my

group on graphene materials and devices. Graphene offers unique properties as a high quality,

“relativistic” 2D electron system that resides right on the surface, can be transferred to arbitrary

substrate, and can be readily probed by a combination of transport, optical and scanning probe

measurements.

Lecture 1 topics: basic material and electronic properties of graphene, graphene fabrication and

synthesis, quantum Hall effect, localization, field effect, disordered graphene, graphene

transistors and graphene sensors;

Lecture 2 topics: advanced and multi-mode experimental techniques: scanning tunneling

microscopy (STM) and scanning gate microscopy (SGM), Raman spectroscopy and imaging,

and in conjunction with device/transport measurements; case studies and applications in

graphene grain boundaries, thermal transport, sensing etc.

For people new to graphene, the following general articles are recommended as preparation for

the lectures: AK Geim and P. Kim, “Carbon wonderland”, Scientific American, April 2008; AK

Geim and NS Novoselov, “The Rise of Graphene”, Nature Materials 6, 183 (2007); AK Geim,

“Graphene: status and prospects”, Science 324, 1530 (2009).

INVITED TALKS

Peter Grutter Director and Fellow, CIFAR Nanoelectronics Program

Physics Department, McGill University

Nanoelectronics – quo vadis?

Thursday 9:00-10:00 am

The 1974 theory paper by Aviram and Ratner, which proposes a molecular rectifier, is generally

credited as starting the field of nanoelectronics. The field seriously started experimentally about

20 years later, culminating in AAAS publication Science declaring „Nanocircuits‟ as the

breakthrough of the year 2001. It was only in 2009, however, that conclusive proof was

presented for transport through one single molecule, as envisaged by Aviram and Ratner [Reed

group, Nature 462, 1039 (2009)] and 2012 for a single atom transistor to be demonstrated

convincingly [Simmons group, Nature Nanotech. 7, 242 (2012)]. Experimental realizations are

finally catching up with the hype of 10 years ago! In this talk I will critically review some of the

challenges and opportunities in nanoelectronics and discuss some of the future directions of this

exciting field.

INVITED TALKS

Andreas Heinrich IBM Almaden Research Center, San Jose, California

The magnetism of atoms and nanostructures on surfaces: an atomic-scale perspective

Thursday 10:30-11:30 am & 1:00 – 2:00 pm

The scanning tunneling microscope has been an extremely successful experimental tool because

of its atomic-scale spatial resolution. In recent years this has been combined with the use of low

temperatures, culminating in precise atom manipulation and spectroscopy with sub-millivolt

energy resolution. In this talk we will demonstrate the possibilities for cutting-edge experiments

that arise from this unique set of “hands and eyes” on the atomic scale.

A sizable cluster of magnetic atoms on a surface behaves similar to a classical magnetic

particle: it‟s magnetization points along an easy-axis direction in space and magnetization

reversal requires sufficient thermal energy to overcome an energy barrier. In this talk we will

discuss how many atoms it takes to create such a bistable magnetic system, which offers crucial

insights into the size limits of magnetic nanoparticles for future applications in data storage and

computation. When the number of magnetic atoms becomes smaller, a smooth transition to

quantum behavior occurs, exemplified by more and more rapid quantum tunneling of

magnetization [1].

Single atoms that are slightly decoupled from conducting substrates behave more like

quantum mechanical entities; classical concepts of magnetization are not appropriate to describe

their behavior. These quantum spin systems can be studied with inelastic tunneling spectroscopy,

a technique we coined spin-excitation spectroscopy a few years back [2]. With this approach it is

possible to measure the energy eigenstates of the quantum spin Hamiltonian that describes spins

on surfaces with high precision. We will introduce its application to the measurement of the

Zeeman energy [2], the magneto-crystalline anisotropy [3], and the spin-spin coupling via a

superexchange interaction [4].

[1] Sebastian Loth, Susanne Baumann, Christopher P. Lutz, D. M. Eigler, and

Andreas J. Heinrich, “Bistability in Atomic-Scale Antiferromagnets”, Science 335, 196

(2012)

[2] A.J. Heinrich, J.A. Gupta, C.P. Lutz, and D.M. Eigler, ”Single-atom spin-flip

spectroscopy”, Science 306, 466 (2004)

[3] C.F. Hirjibehedin, C.-Y. Lin, A.F. Otte, M. Ternes, C.P. Lutz, B.A. Jones, A.J. Heinrich,

"Large Magnetic Anisotropy of a Single Atomic Spin Embedded in a Surface Molecular

Network", Science 317, 1199 (2007)

[4] C.F. Hirjibehedin, C.P. Lutz, A.J. Heinrich, “Spin-coupling in engineered atomic

structures”, Science 312, 1021 (2006)

INVITED TALKS

Werner Hofer Stephenson Institute for Renewable Energy, The University of Liverpool

Unraveling electron mysteries: data analysis in scanning tunneling microscopes

Thursday 2:30 – 3:30 pm

The lecture starts by explaining, how physical processes can change the measured values in a

scanning tunneling microscope (STM), demonstrating with a few examples that the relation

between atomic arrangement, electronic properties, and measured currents and spectra is far from

trivial. On this basis it is argued that only quantitative theoretical predictions are sensible,

because only they allow an estimate about the quality of comparisons between experimental and

theoretical results. This theme is developed in particular in magnetic STM, where different STM

tip models lead to completely different theoretical predictions. This is also demonstrated for the

measurement on metal surfaces, where it is shown that qualitative results are only in accordance

with experimental data, if the distance range is severely underestimated.

Dynamic processes observed by scanning tunneling microscopes:

vibrations, diffusions and reactions

Friday 9:00 – 10:00 am

Dynamic processes in scanning tunneling microscopy (STM) are increasingly the focus of

cutting edge research due to their importance for energy conversion and reaction processes. It is

in principle possible to study these processes by suitable adaptation of STM theory and a step-

by-step analysis of the processes themselves. I shall give several examples where such a detailed

understanding is indispensible for a comprehensive understanding e.g. in atomic switching and

diffusion processes, in molecular growth processes, condensation reactions, and long range

molecular propagation even on reactive surfaces. At the end of my talk I shall demonstrate that

careful statistical analysis in combination with high-resolution STM can even lead to surprising

new insights into fundamental physics.

STUDENT ORAL PRESENTATIONS

Electronic structure of graphene nanoribbons with metallic contacts Thursday 3:30 – 3:50 pm

Chloé Archambault, Alain Rochefort

École Polytechnique de Montréal, C.P. 6079, Succ. Centre-ville, Montréal, Québec, H3C 3A7

E-mail: [email protected]

Graphene is a very promising material for electronics due to its extremely high electron mobility.

Its two-dimensional structure is especially well suited for the current microfabrication

techniques. It can be made semiconducting after being engineered into nanoribbons [1, 2] such

that graphene transistors have already been reported [3, 4].

The electronic interactions between graphene nanoribbons (GNRs) and the indispensable

metallic contacts remains a source of high interest since critical features at such a small length

scale can have a dramatic impact on the final device performance. For example, metal induced

gap states (MIGS), which have been observed experimentally [5], could play an important role in

transport at low bias. A better understanding of these effects is needed for further device

optimization.

Here we report the results of first principles density functional theory (DFT) calculations of finite

nanometer-size GNRs contacted with Au and Pd electrodes. We observe significant charge

transfer from graphene to the electrode which extends into the nanoribbon beyond the zone

located immediately under the contact. In the case of zigzag GNRs, the charge transfer is mainly

accommodated by the ribbon's edge states. The local density of states (LDOS) at the Fermi level

reveals a long range interaction penetrating deep into the ribbon which is explained by the effect

of the metal induced charge transfer on the GNRs' frontier orbitals. For Pd, there is clear

evidence of the hybridization between the GNR's frontier orbitals and the metallic states, giving

rise to MIGS that decay in the GNR about 1 nm away from the contact.

MIGS in a zigzag GNR contacted with Pd

[1] K. Nakada, M. Fujita, G. Dresselhaus, and M. Dresselhaus, Phys. Rev. B 54, 17954 (1996)

[2] M. Han, B. Özyilmaz, Y. Zhang, and P. Kim, Phys. Rev. Lett. 98, 206805 (2007)

[3] X. Wang, Y. Ouyang, X. Li et al., Phys. Rev. Lett. 100, 206803 (2008)

[4] Z. Chen, Y.-M. Lin, M. J. Rooks, and P. Avouris, Physica E 40, 228 (2007)

[5] F. Xia, T. Mueller, R. Golizadeh-Mojarad et al., Nano Lett. 9, 1039 (2009)


Solution of the Contact Problem of Molecular Nanoelectronics Thursday 3:50 – 4:10 pm

Firuz Demir, George Kirczenow1

Simon Fraser University


Molecular nanowires in which a single molecule bonds chemically to two metal electrodes and

forms a stable electrically conducting bridge between them have been studied intensively for

more than a decade. However the experimental determination of the bonding geometry between

the molecule and electrodes has remained elusive. In this talk it will be shown how inelastic

tunneling spectroscopy (IETS) experiments when combined with theory can identify the atomic

scale structures of the molecule-metal interfaces of single-molecule nanowires bridging pairs

metal electrodes, and thus can resolve the long- standing “contact problem” of molecular

nanoelectronics. We consider the propanedithiol (PDT) molecules bridging gold nanocontacts in

the recent experiment of Hihath et al. [Nano Lett. 8, 1673 (2008)]. Based on ab initio density

functional and semi-empirical calculations, we find the relaxed geometries and vibrational modes

of extended molecules each consisting of one or two PDT molecules connecting two gold

nanoclusters and calculate their inelastic tunneling spectra. Comparing our results with the data

of Hihath et al., we find that the most frequently realized conformation in the experiment was

trans molecules top-site bonded to both electrodes. We identify the features observed in the

experimental IETS of these molecules at phonon energies near 46, 40 and 42 meV and show the

switching from the 42 meV vibrational mode to the 46 meV mode observed in the experiment to

be due to the transition of trans molecules from mixed top-bridge to pure top-site bonding

geometries. PDT molecules in which the sulfur atoms retain their thiol hydrogen atoms and bond

strongly to gold in the top site geometry give rise to an IETS feature in the phonon energy range

54-57 meV. For pairs of PDT molecules connecting the gold electrodes in parallel we find total

elastic conductances close to twice those of single molecules bridging the contacts with similar

bonding conformations and small splittings of the vibrational mode energies for the modes that

are the most sensitive to the molecule-electrode bonding geometries.

Top and bridge site bonding of trans PDT molecules

[1] F. Demir, G. Kirczenow, J. Chem. Phys. 136, 014703 (2012)


Magnetoresistance in Hydrogenated Graphene Thursday 4:10 – 4:30 pm

Jonathan Guillemette1, Shadi Sabri

1, Binxin Wu

1, Pierre Lévesque

2, Abdelaadim Guermoune

3,

Mohammed Siaj3, Richard Martel

2, Guillaume Gervais

1, Thomas Szkopek

1

1McGill University, Physics Department

2École Polytechnique, Départment de Génie Chimique

3Université du Québec à Montréal, Départment de Chimie


Using a custom built hydrogenation gun in a UHV system, graphene samples hydrogenated to

various degrees were prepared. Room temperature resistance increases ranging from a few

percent to a 3 order of magnitude increase were achieved. The analysis of the thermal transport

of the hydrogenated graphene samples has revealed a metal insulator transition. By inserting

these samples in a 0.3K 9T magnetic field, negative magnetoresistance all the way to 9T was

found. By using the Tallahassee Magnetic Laboratory, we were able to have access to a 30T

magnet which displayed negative magnetoresistance all the way up to those magnetic fields. This

is the first report of negative magnetoresistance going all the way to 30T in hydrogenated

graphene. This disordered system can be used to explore the limits of weak and strong

localization as well as the metal – insulator transition. The investigation of the evolution of

scattering times as a function of the Fermi level of the hydrogenated graphene will be presented.

1- Variable range hopping in hydrogenated graphene

2- Normalized magnetoresistance trace at various temperatures

[1] X. Hong, S. H. Cheng, C. Herding, J. Zhu, Physical Review B 83, 085410 (2011)

[2] J. Moser et al., Physical Review B 81, 205445 (2010)

[3] Y. F. Chen et al., Journal of Physics-Condensed Matter 22, 205301 (2010)


Effect of alkyl chain-length on dissociative attachment in two families of

reaction at Si(100) Friday 10:00 – 10:20 am

Maryam Ebrahimi1, Si Yue Guo, Kai Huang, Tingbin Lim, Iain R. McNab

2, Zhanyu Ning, John

C. Polanyi, Mark Shapero, and Jody (S.Y.) Yang

Department of Chemistry, University of Toronto, Toronto, ON, Canada 1Department of Chemistry, University of California, Riverside, Riverside, CA, USA

2School of Biological Sciences & Applied Chemistry, Seneca College, Toronto, ON, Canada

Presenting author E-mail: [email protected]

Corresponding author E-mail: [email protected]

We studied the dissociative attachment (D.A.), on silicon, of two families of molecules:

1-bromoalkanes and primary alcohols (which we will refer to as R-Br and R-OH, where R is a

carbon chain). The question of interest is the effect of increasing chain length, R, from C2H5 to

C5H11, on the chemical reactivity at the Si(100) surface. Each molecule was examined by

scanning tunneling microscopy (STM) at temperatures ranging from 50 K to 300 K to determine

(i) the geometry of species at the surface, (ii) the pattern of reaction to give D.A., and (iii) the

activation energy for D.A. The results were interpreted by ab initio quantum mechanical

calculation.

The three 1-bromoalkane molecules studied, bromoethane (EtBr), 1-bromopropane (PrBr), and

1-bromobutane (BuBr), were seen to physisorb and react exclusively in an inter-row

configuration on the Si(100)-c(4x2) with activation energies Ea of 0.343±0.005 eV for EtBr,

0.410±0.006 eV for PrBr, and 0.536±0.002 eV for BuBr [1]. In constrast, the three primary

alcohols, ethanol (EtOH), 1-propanol (PrOH), and 1-pentanol (PeOH), reacted exclusively on

inter-dimer sites. The energy barriers were measured to be 0.41±0.02 eV for EtOH, 0.60±0.01

eV for PrOH, and 1.004±0.007 eV for PeOH. The molecules in both families, R-Br and R-OH,

were found to have increasing activation energy for every additional carbon in the alkyl-chain.

Extensive calculations using density function theory was completed for R-Br, confirming the

increase in energy barrier by amounts comparable with the observed values. We interpret the

increase in Ea in terms of the additional energy required to lift the alkyl-chain away from the

surface, going from the initial physisorbed state to the transition state. Calculations are in

progress for R-OH.

[1] Ebrahimi, M. et al, J. Phys. Chem. C, in press (web publication on April 12, 2012)


Limits of Detection for Silicon Nanowire BioFETs Friday 10:20 – 10:40 am

Nitin K. Rajan1, Xuexin Duan

2, Aleksandar Vacic

2, David A. Routenberg

2, Mark A. Reed

1,2

1Dept. of Applied Physics, Yale University, 15 Prospect St, New Haven CT06511

2Dept. of Electrical Engineering, Yale University, 15 Prospect St, New Haven CT06511


Over the past decade, silicon nanowire/nanoribbon field-effect transistors (NWFETs) have

demonstrated great sensitivity to the detection of biomolecular species, with limits of detection

(LOD) down to femtomolar concentrations [1]. Several well known factors limit the LOD;

among them, ionic concentration, efficiency of the biomolecule-specific surface

functionalization, binding constants, and the delivery of the analyte to the sensor surface.

However, the signal-to-noise ratio (SNR) of these bioFET sensors, and the device parameters

that determine the LOD, are not well understood. In this work, we apply noise spectroscopy (Fig

1.) to silicon NWFETs with the goal of understanding and improving the detection limit of such

devices. Using low frequency noise measurements, we quantify the effect on device performance

of different process parameters. We also show that the SNR is maximized at maximum

transconductance (Fig 2.) due to the effects of 1/f noise, and not in the subthreshold regime

where sensitivity is maximized [2]. These devices currently have a LOD of 4 electronic charges

in ambient conditions. Using these devices, with very good performance in terms of SNR, we

were able to measure and extract the binding kinetics of protein interactions, which have never

been done with NWFETs. Binding constant determination is a critical parameter for

biomolecular design, and has until now been primarily assessed by surface plasmon resonance

(SPR). Utilizing the low LOD of these devices, we are able to extract binding constants into the

sub-picomolar range.

Fig. 1 Typical Noise Spectra for field effect

biosensors as a function of solution gate voltage.

-1.0 -0.8 -0.6 -0.4 -0.2 0.0

20000

25000

30000

35000

40000

45000

50000

55000

60000

SNR

Gm

Solution Gate Voltage Vsg

(V)

SN

R (

V-1)

0

200

400

600

800

1000

1200

gm (n

A/V

)

Fig 2. Signal-to-noise ratio plotted as a function of

solution gate voltage. Peak SNR occurs at -0.5V and

is equal to 59000 V-1

. Transconductance (gm) also

plotted as a function of solution gate bias confirming

that SNR is maximized close to the peak gm.

[1] Stern et al. Nature 445, 519-522 (2007)

[2] Rajan et al. Applied Physics Letters 97, 243501 (2010)

STUDENT POSTER PRESENTATIONS – POSTER 1

b)

A Highly Sensitive Au-string Nanoresonator as a Sensor

T.S. Biswas

1, A. Suhel

1, B.D. Hauer

1, K.S.D. Beach

1 and J.P. Davis

1, 2

1Department of Physics, University of Alberta, Edmonton, Alberta, Canada T6G 2E9

2Canadian Institute for Advanced Research, Toronto, Ontario M5G 1Z8, Canada


Two important aspects of nanomechanical devices for physics applications are high quality

factor (Q) and an easy, yet sensitive, measurement technique. The Q is a measure of the stored

energy in a mechanical resonator. A higher Q means that a resonator stores more energy,

resulting in a probe with better sensitivity. We have made nanoresonators from high stress

silicon nitride shown in Fig. 1a, called nanostrings, for which we have measured Qs on the order

of 105 to 10

6 using optical interferometry shown in Fig. 2. [1] In addition, through

thermomechanical calibration and comparison with theory, we have learned that the major

contribution to dissipation in our nanostrings comes from clamping loss at the boundaries -

demonstrating a way to further improve our devices. We are also working to simplify our

detection scheme to read out the nanostrings. Our goal is to perform all-electrical measurements

using the piezoresistance of gold that has been coated on the nanostrings. [2] We have

successfully fabricated devices using ~40 nm gold on the top of high stress silicon nitride shown

in Fig. 1b and have measured Qs on the order of 105, confirming that there is no significant

change in Q from nanostrings without gold. Finally, we will report our progress with the

electrical measurements of gold-coated nanostrings. When successful, we believe our devices

will enable a variety of experiments in extreme environments where laser readout is difficult,

like low temperatures.

[1] A. Suhel, B.D. Hauer, T.S. Biswas, K.S.D. Beach and J.P. Davis, APL, In press, 2012

[2] M.L. Roukes, arXiv:cond-mat/008187v2

a)

Fig. 2. Qs for different modes of Si3N4

nanostring (215 m long)

Fig. 1. Scanning electron micrographs of a) Si3N4

string and b) Au on the top of Si3N4

20 m


Graphene Nanomechanical Resonators

Arnab Chaudhuri1, R. Saydur Rahman

2, R. G. Knobel

3

1PhD Candidate, Queen‟s University, Canada

2Post Doctoral Fellow, Memorial University, Canada

3Associate Professor, Queen‟s University, Canada


Graphene, a single atomic layer of graphite, possesses remarkable properties has drawn attention

of physicists, chemists and engineers since its extraction in 2004 [1]. Along with its fascinating

electronic properties [2], graphene because of its high Young‟s modulus, low mass, high surface

area and high quality factor (Q) (Q ~ 100000 at low temperature) [3] is an ideal candidate for

sensitive mechanical resonators. Graphene is extracted using mechanical exfoliation [4] or

chemical vapour deposition [5] transferred to Si-Si oxide substrate and then suspended graphene

device is fabricated using electron beam lithography, metal deposition to build electrodes and

substrate etching [6]. In this poster we present our progress towards realizing graphene

nanomechanical resonators for both fundamental experiments and applications as sensors. The

device is driven into vibrational motion by applying an oscillating voltage to the graphene bridge

and can be controlled by varying the back gate voltage. The readout is detected electrically by

measuring the output current through the graphene flake, which is a function of vibrational

frequency and graphene-substrate distance, using a lock in amplifier. If the temperature of the

system is lowered, the vibrational motion will tend towards quantum zero-point motion, which

will be manifested with increased visibility for graphene resonators, owing to their extremely

light mass. Tuning the driving signal introduces a duffing non-linearity, which modulates the Q

factor thus providing control over its mechanical properties [7]. Apart from these studies an

immediate application of these devices will be for ultra-sensitive mass and force detection.

[1] Novoselov et. al., Science 306, 666 (2004)

[2] Castro Neto et. al., Rev. Mod. Phys. 81, 109 (2009)

[3] Chen et. al., Nature Nanotechnology 4, 861 (2009)

[4] Geim et. al., Nature Materials 6, 183 (2007)

[5] Reina et. al., Nano Letters 9, 30-35 (2009)

[6] Bunch et al., Science 315, 490-493 (2007)

[7] Eichler et. al., Nature Nanotechnology 6, 339 (2011)


Thermal transport in partially unzipped carbon nanotubes

Xiaobin Chen,1 Yong Xu,

1 Bing-Lin Gu,

1 Wenhui Duan

1*

1Department of Physics, Tsinghua University, Beijing 100084, China

*E-mail: [email protected]

We report systematic investigation on the thermal properties of partially unzipped carbon

nanotubes (PUNTs) using the nonequilibrium Green's function method. It is found that thermal

conductance shows linear dependence on the central unzipped width and exponentially decays to

a finite value as the unzipped part increases. Further analysis on one dimensional atom chains

confirms that the exponential decay is due to interface scattering of the mismatched modes. Our

findings of wide-range linear controllable thermal conductance in PUNTs may help find high-

performance thermoelectric devices and the exponential decay behaviour may lead to a better

understanding of phonon transport in nanoscale systems.


Investigation of kinetics and growth mechanisms of graphene on copper

Saman Choubak, Pierre Levesque, Patrick Desjardins, Richard Martel

École Polytechnique, Départment de Génie Chimique


Graphene, a single atomic layer of sp2-bonded carbon atoms packed to a two-dimensional

honeycomb lattice, has attracted scientists since its discovery . It is potentially a next generation

material for the electronic industry due to its novel physical and electrical properties. Graphene‟s

high carrier mobility and phenomenal optical transparency makes it an ideal material for

numerous applications in electronics, optoelectronics, biosensors, etc. In order for graphene to

become useful in practical applications, its uniform and large-area mass-production with low

defects and controlled thicknesses has become crucial.

In order to shed light to critical aspects of graphene growth mechanisms and kinetics, we are

using low energy electron microscope, (LEEM), to observe and investigate graphene nucleation

and growth mechanisms on single crystal copper. The ultimate purpose of this project is to

elucidate a viable and reproducible avenue for the production of high quality and

monocrystalline graphene films. In situ LEEM observation provides information of the processes

taking place on the surface, which allows a thorough description and understanding of growth

features. In our experiments, LEEM will serve as a tool to study graphene growth on single

crystal copper exposed to a flux of elemental carbon, atomic hydrogen, and C-based precursors.

The design of the experiments is aimed to imitate the surface kinetics involved in CVD growth

of graphene on copper foils and to gain insight into the growth mechanisms. These observations

demonstrate graphene growth from vapor-deposited carbon on the surface and the phenomenon

that occur in order for the film to nucleate and grow . This knowledge and information will allow

the control of the process in detail, whereas, otherwise stable structures may not be produced

with perfection.


Theory of anomalous magnetotransport in triple quantum dots

B. D’Anjou, W.A. Coish

McGill University, 3600 rue University, Montreal (QC), Canada H3A 2T8


Magneto-transport measurements on a triple quantum dot ring have recently shown anomalous

quantum oscillations with dominant frequencies separated by a factor of three in magnetic flux

[1]. Such oscillations, suggestive of a one-third periodicity in the flux quantum, are usually not

observed in larger mesoscopic rings in which only larger periods are observed. We develop a

microscopic transport model for the triple dot and show that the anomalous oscillations can

dominate the transport behavior under certain conditions. Furthermore, we discuss the range of

validity of our model by studying dephasing due to broadening and electric dipole interactions.

[1] L. Gaudreau et al., Phys. Rev. B 80, 075415 (2009)


Structural profiling and electronic properties of pentacene on Ni(111) surface

Laurentiu Eugeniu Dinca, Jennifer MacLeod, Csaba E. Szakacs, Josh Lipton-Duffin,

Dongling Ma and Federico Rosei

INRS-EMT, 1650 boulevard Lionel-Boulet, Varennes, QC, J3X 1S2, Canada


Using scanning tunnelling microscopy (STM), we have investigated the adsorption of pentacene

on Ni(111) surface. Further insight into the ordering and interaction mechanism of pentacene on

the metal surface is revealed by theoretical simulations, based on density functional theory

(DFT). At room temperature and close to monolayer coverage a random structure of pentacene

emerges (Figure 2). Pentacene molecules are evenly distributed with their longitudinal axes

along the Ni(111) high-symmetry directions. A new structure reveals after annealing to 200 0C,

which we interpret as two pentacene blocks π-π stacked or covalently attached to each other in

graphene or peripentacene (Figure 3). Prolonged annealing of sub-monolayer coverage produces

pentacene dissociation. We have also investigated the energetics of the adsorption sites of

pentacene at monolayer coverage for different single molecule-substrate configurations using

DFT.

Figure 2. STM images of pentacene molecules adsorbed on Ni(111) substrate, no pre-annealing. (a). 30nm × 30nm, I

= 1.36 nA, U = 1.83 mV, and (b). 3nm × 3nm, I = 1.49 nA, U = 1.83 mV.

Figure 3. STM images of pentacene packing structures on Ni(111) substrate, pre-annealed to 200ºC for 15 minutes.

(a). 10nm × 10nm, I = 1.47 nA, U = 10.68 mV, and (b). 3nm × 3nm, I = 1.51 nA, U = 45.47 mV.


Étude comparative du taux de recristallisation interfacique entre le silicium

amorphe relaxé et le silicium amorphe dé-relaxé

Ousseynou Diop, Sjoerd Roorda

Département de physique, Université de Montréal, Montréal (Québec), H3C 3J7, Canada


Nous avons montré une différence entre les taux de recristallisation interfacique pour le silicium

amorphe relaxé versus celui dé-relaxé. Pour cela nous avons amorphisé deux zones distinctes sur

une gaufre de silicium cristallin en lui bombardant par des ions 28Si+. C‟est à dire on effectue

une série d‟implantations sur le c-Si avec des énergies de 2 MeV, 1 MeV et 0.5 MeV

consécutives à une fluence de 5x1015 ions/cm² chacune. Après un recuit, RTA (Rapide Thermal

Annealing) de relaxation pendant 5s à 650 °C dans un four et réimplantation d‟une zone, les

deux échantillons ont une épaisseur amorphe de (1.50±0.07) μm. La recristallisation débute à

l‟interface et s‟étend graduellement à une certaine vitesse vers la partie amorphe avec le nombre

de recuits. Une série de recuit à (500±4) °C, nous a permis de montrer que la différence entre la

zone relaxée versus dé-relaxée est significative après 4 heures de recuit et est de (0.096 ±0.007)

μm, ce qui correspond à 4.8x1017

atomes déplacés par cm2. Les taux de recristallisations sont à

l‟ordre de (2.722±0.005)×10-9

cm/s et (1.826±0.003)×10-9

cm/s respectivement dans l‟échantillon

dé-relaxé et relaxé. Soit environ 1.490±0.004 fois plus important dans l‟état dérelaxé. Ces

valeurs diminuent graduellement avec le nombre de recuits. Nous avons constaté que cette

différence de taux de recristallisation est liée à la réduction de densité de défauts supplémentaires

produits dans le silicium amorphe dérelaxé. Après le septième recuit cette différence diminue

considérablement. Ce qui veut dire que nos deux échantillons se recristallisent progressivement

avec le même taux à une légère différence de (0.14 ±0.07) x10-9

cm/s pour l‟échantillon relaxé.


Phase stack limit charge transport in polymer photovoltaic devices

Fei Dou1,2

, Carlos Silva2, Baozeng Wang

1, Le Kong

1, Xinping Zhang

1

1Institute of Information Photonics Technology and College of Applied Sciences,

Beijing University of Technology, Beijing 100124, China 2Département de Physique and Regroupement Québécois sur les Matériaux de Pointe, Université

de Montréal, Montréal (Québec), H3C 3J7, Canada


We found the external quantum efficiency of the photovoltaic device of poly(9,9¢-

dioctylfluorene-co-bis(N,N¢- (4,butylphenyl) bis(N,N¢- phenyl -1,4- phenylene)diamine) (PFB)

and poly(9,9¢-dioctylfluorene- co-benzothiadiazole) (F8BT) can be strongly modified through

changing the mixed polymer composite ratio, which obtained the optimum at ratio 1:3 of F8BT:

PFB. After comparing the influence of charge separation and charge recombination at polymer

interface, and comparing different charge transport road at different composite ratio samples, we

proposed that not only the phase interface area, but also the phase stack configuration can largely

influence the device performance. It implies that the device performance can be improved by

better control the phase separation and phase stack through changing the blend film morphology.


Instrumentation and Methods for Inertial Motors used in Scanning Probe

Microscopes

Benedict Drevniok1, William M.P. Paul

2, Alastair B. McLean

1

1Department of Physics, Queen's University, Kingston, Ontario, Canada

2Department of Physics, McGill University, Montreal, QC, Canada


Scanning probe microscopy continues to evolve and push the boundaries of research into basic

electronic and mechanical behaviour of surfaces and adsorbates. The ultra-high vacuum, low-

temperature, high magnetic field environments that are becoming more common in fundamental

SPM research place a unique set of demands on the instrumentation used to approach and scan a

sample. As many of the highest-performance instruments remain home-built, it is important for

wide-spread understanding of the basic building blocks of microscope operation.

We present information that will facilitate the assembly, characterization and operation of

inertial motors for modern scanning probes [1]. Specifically, a model describing the forces

required to create a step in a stick-slip type motor is developed. Simple, inexpensive drive

electronics with high slew rate and high output current are described that have been used with a

number of commercial controllers. Additionally, a novel reflective object sensor that can be used

to characterize motor performance is described. We illustrate the application of the drive

electronics and the optical sensor using a Pan-style inertial motor [2] that we have used in our

laboratory for the last three years.

[1] B. Drevniok et al., Rev. Sci. Instrum. 83, 033706 (2012)

[2] S.H. Pan et al., Rev. Sci. Instrum. 70, 1459 (1999)


Design and fabrication of QCA based on a SET process

G. Droulers1, S. Ecoffey

1, M. Pioro-Ladrière

2, D. Drouin

1

13IT, Université de Sherbrooke, Sherbrooke, Quebec J1K 2R1, Canada

2Départment de physique, Université de Sherbrooke, Sherbrooke, Quebec J1K 2R1, Canada


As the MOSFET is further scaled down, the increase in undesirable quantum effects and the need

for better energy efficiency is driving the development of new technologies and computational

methods. The quantum cellular automata (QCA) paradigm, where two excess electrons are

confined in two of four quantum dots placed at the corners of a square, was introduced

in 1993 [1]. Since then, many theoretical and experimental demonstrations have been made.

However, room temperature operation is still a challenge especially if one wants to integrate

QCA to usable, everyday life technologies. Here, we propose a technology platform for QCA

room temperature operation. Our platform is based on the nanodamascene process [2], used for

the fabrication of single electron transistor (SET).

The first part of the project concerns the design and simulation of a clocked QCA half-cell with

SET readout (see [3]). This involves finite element method for capacitance extraction and

numerical solving of the Coulomb blockade master equations. The clocking is achieved by the

addition of a third, centered dot in the half-cell to enable a “NULL” state. The second part of the

project is to fabricate and characterize devices and demonstrate the application of the process to

these QCA principles. Intermediate devices like nanowires and metal-insulator-metal (MIM)

capacitors have been fabricated and characterized, but metal-oxide interface states remain a

challenge. Low temperature electrical characterization and devices optimisation will follow in an

attempt to raise the operating temperature. The new paradigm introduced with QCA in

combination with the present process has the potential to greatly reduce the power consumption

of future electronics and this project is a step in that direction.

Schematic of the process used (a) with fabrication design (b) and equivalent circuit (c). Simulated operation of the

circuit (d) and scanning electron microscopy image of a sample midway through the process (e).

[1] A. Beaumont, C. Dubuc, J. Beauvais et al., IEEE Electr Device L 30, 766 (2009)

[2] C. S. Lent, P. D. Tougaw, W. Porod et al., Nanotechnology 4, 49 (1993)

[3] A. O. Orlov, I. Amlani, R. Kummamuru et al., Appl Phys Lett 77, 295 (2000)


Encapsulated graphene field effect transistors

S A Imam, A Guermoune, M Siaj, T Szkopek

Department of Electrical and Computer Engineering, Department of Physics, McGill University


Graphene field effect transistors are fabricated on silicon oxide and silicon nitride substrates.

They are then encapsulated by 10nm thin silicon oxide and silicon nitride by room temperature

magnetron sputtering. The oxide encapsulated devices retain 55% of its bare device mobility,

while nitride retains almost 80%. Both the films result in minimal lattice degradation of

graphene, as verified by Raman Spectroscopy.


Characterization of Silicon Nanoribbon Debye Length Modulation using DNA

M. H. Izadi1, J. Thomas

2, X. Duan

3, D. Sen

2, P. Grutter

1, M. Reed

3

1Dept. of Physics, McGill University, Montreal, Canada

2Dept. of Molecular Biology and Biochemistry, Simon Fraser University, Burnaby, Canada

3Depts. of Elec. Eng. and Applied Physics, Yale University, New Haven, CT, USA


Silicon nanoribbon field effect transistors (SiNR-FET) have the ability to sense changes in the

charge distribution of surface bound biomolecules [1] which gives them the potential to monitor

molecular charge distributions of chemical reactions in real-time. We employ DNA in order to

explore the capabilities of SiNR-FET charge detection from bound biomolecules.

SiNR-FETs can sense the molecular charge distribution of bound biomolecules with changing

ion concentration of the buffer solution. A changing ion concentration will change the Debye

screening length and thus change how much of the charge associated with the bound

biomolecules the nanoribbon „sees‟ (Figure 1).

We use duplex and triplex DNA to serve as „charge rulers‟ due to their regularly and finely

spaced negative charges along the DNA backbone (3.4 Angstroms between charges). Proper

formation of triplex DNA is confirmed using circular dichroism spectroscopy. Attachment of

DNA to silicon surfaces has been characterized using both fluorescent tagging and XPS. Initial

studies indicate effective charge screening modulation with varying salt buffer concentration.

Figure 1. SiNR-FET with duplex DNA attached. Increasing the salt buffer concentration reduces the Debye

screening length and hence the amount of attached biomolecule charge the nanoribbon „sees‟.

[1] A. Vacic, J. Criscione, N. K. Rajan, et al., J. Am. Chem. Soc. 133, 35 (2011)


Raising the bar on optomechanics in Davis lab

Paul H. Kim1, Bradley D. Hauer

1, Callum Doolin

1, and John P. Davis

1,2

1Department of Physics, University of Alberta, Edmonton, Alberta, Canada T6G 2G7

2Canadian Institute for Advanced Research, Toronto, Ontario, Canada M5G 1Z8


Recently, there has been a breakthrough in measuring mechanical displacements of on-chip

nanomechanical devices, which has been named optomechanics [1]. This ultra-sensitive

measurement technique goes beyond the traditional interferometric scheme and is made possible

through a strong correlation between an optical cavity and a mechanical resonator. In essence, a

high quality factor for the optical cavity yields increased sensitivity for nanomechanical

detection. From advances in fabrication methods, ultra-high quality factors on the order of ~106

have been achieved using various types of Fabry-Perot, whispering gallery, and photonic cavities

[2]. Our ultimate goal is to further investigate quantum properties of nanoscale devices at low

temperatures. Our lab is currently under construction, installing two nuclear demagnetization

dilution refridgerators that will achieve lowest temperatures in Canada. After completion, I will

test various designs of nanomechanical devices, which are extensions of previously fabricated

devices in our lab [3]. We highlight our current progress towards fabricating a successful

dimpled tapered fiber, a custom machined vacuum chamber for environment control, and the

design of silicon-on-insulator optomechanical devices.

Optical image of the fabricated dimpled fiber, SEM image of the 80 m disk with a mechanical beam, and the

vacuum chamber for the apparatus.

[1] G. Anetsberger et al., arXiv:1108.4608v1, (2009)

[2] K. J. Vahala., Nature 424, 6950 (2003)

[3] J.P. Davis, et al., Appl. Phys. Lett. 96, 072513 (2010)


Graphene Studied by Highly Coherent Low Energy Electron Holography

A. Peter Legg, Josh Mutus, Lucian Livadaru, Radovan Urban, Jason Pitters, Robert A. Wolkow

Department of Physics, University of Alberta, 11322 89 Ave , Edmonton Alberta 1National Institute of Nanotechnology, 11421 Saskatchewan Drive, Edmonton Alberta


Advances in low energy (~100 eV) in-line point projection microscopy had stalled at the ~1 nm

scale as a consequence of nano-tip emission limitations and possibly also as a result of

instrument instabilities. The recently developed nitrogen etched and stabilized tungsten nano-tip,

together with lowered mechanical noise and reduced stray magnetic perturbations have allowed a

>4 times increase in coherence angle and a virtual source size of 1.6 +/- 0.6 Angstrom, evidently

opening the door to atom-scale holographic imaging of molecules and other nano-scale entities.

Point projection microscopy will allow us to probe a graphene surface for defects as previous

experiments have shown no adverse effects from imaging. Theoretically, 3 dimensional

reconstructions of the surface are also possible which may reveal important properties invisible

to other imaging techniques.

Starting with Gabor, and by touching upon the work of other contributors to the field, the idea

and practice of point projection microscopy will be briefly reviewed. Our particular, ultra high

vacuum, instrument will be described. The instrument is at once both an atom-scale imaging

scanning tunneling microscope and a point projection microscope. The stable, reproducible and

readily repairable nitrogen etched nano-tip will be described. Results of a published study of

graphene, including observation of opacity changes with heat cleaning, and observations of ~nm

scale corrugation of suspended graphene will be shown [1]. Characterization of the electron

beam will be shown. Un-supported graphene/vacuum knife edge studies, among others, will be

described in the context of beam coherence properties. Where in the past a maximum coherence

half angle of 2 or 3 degrees had been observed [2], with the new apparatus a half coherence

angle of over 14 degrees is obtained. Some discussion of the interwoven properties of lateral and

longitudinal coherence, together with energy spread will be offered.

Graphene edge in point projection microscopy

[1] J. Y. Mutus, L. Livadaru, J. T. Robinson et al., New J. Phys. 13, 063011 (2011)

[2] Che-Cheng Chang, Hong-Shi Kuo et al., Nanotechnology, 20, 115401 (2009)


Ising spins on a tuneable fcc to bcc lattice

Andrew Macdonald, Sarah Burke, Doug Bonn, Yan Pennec

Depart. of Physics and Astronomy, University of British Columbia, V6T1Z4 Vancouver, Canada


The study of magnetically frustrated materials has led to the discovery and control of novel

nano-scale devices [1]. A prime example is the phenomenon of exchange bias, which caused a

revolution in hard-drive read-head technology. Current research hints at further breakthroughs,

from experimental realization of ac magnetic “currents” in spin ice materials [2] to construction

of a quantum computer using magnetic qubits [3]. Future progress in frustrated magnetism will

require an strong union of experiment and theory. It is in this spirit that we report the results of a

Monte Carlo study of an Ising spin system on a tunable fcc to bcc lattice aimed at describing the

magnetic configuration of thin Cr films grown epitaxially on Au(100).

Cr has been studied extensively as the “archetypical itinerant antiferromagnet” [4]. In the early

1990s the surface of Cr(100) was the first system to be studied with spin-polarized scanning

tunneling microscopy (SP-STM), which demonstrated “topological antiferromagnetic order”

between the magnetic moments in neighbouring atomic planes [5]. In recent work at UBC it was

found that thin film fcc Cr(100) can be grown epitaxially on Au(100). An antiferromagnet on a

fcc lattice is strongly geometrically frustrated and SP-STM images of the fcc Cr(100) surface

showed that this frustration fundamentally altered the magnetic order. The surface exhibited a

complex magnetic topography with regions of ferromagnetic and antiferromagnetic order

separated by domain walls within the same [100] plane. These ordered regions were contrasted

by a breakdown of the topological order seen in bulk bcc Cr, with no apparent correlation

between spin directions in adjacent atomic terraces. The observed spin structure did not

correspond to that of a pure bcc or fcc lattice but a structure containing elements of both lattices.

To model the influence of frustration in this system computationally we employed a classical

Monte Carlo simulation of Ising pseudo-spins with nearest neighbour antiferromagnetic

exchange and single spin flip dynamics. Simulating 105 to 10

6 spins on a tuneable fcc to bcc

lattice we let the dynamics evolve with the Metropolis algorithm and examined the evolution of

the ground state as a function of the tuning parameter and simulation cell boundary conditions. A

comparison of the simulation with experiment shows strong agreement, and explains how the

inherent geometric frustration of the fcc lattice causes the breakdown of topological order

between successive atomic planes.

[1] J. P. Liu et al., Nanoscale Magnetic Materials and Applications, Springer (2009)

[2] O. Tchernyskyov, Nature New & Views 451, 22 (2008)

[3] J. Tejada, E. M. Chudnovsky, E del Barco et al. Nanotechnology 12, 181 (2001)

[4] E. Fawcett, Rev. Mod. Phys. 60, 209 (1988)

[5] R. Wiesendanger, H. J. Güntherodt et al., Phys. Rev. Lett. 65, 247 (1990)


Dynamics of one and two-bond dissociations of 1,2-dihaloethanes on Si(100)

Oliver MacLean, Si Yue Guo, Kai Huang, K. R. Harikumar, Amir Zabet-Khosousi, Iain

McNab1, John C. Polanyi

Department of Chemistry, University of Toronto, 80 St. George Street, Toronto, ON M5S 3H6 1Biol. Sciences and Appl. Chemistry, Seneca College, 70 The Pond Rd, Toronto, ON, M3J 3M6

Corresponding author e-mail: [email protected]

Colleagues in my group recently observed that 1,2-dichloroethane and 1,2-dibromoethane

dissociate on Si(100) in intriguingly different ways at room temperature and 110 K. At room

temperature, both carbon-halide bonds dissociate to form ethylene and a pair of halogen atoms,

with the halogen atoms bonded to Si atoms on the same dimer [1]. The ethylene is found on

average about 30 Å away from the halogen pair, which was attributed to “cartwheeling” of the

hot ethylene across the surface.

We examined the dissociation at lower temperature to investigate what further dynamics could

be in play. Surprisingly, at 110 K, only one carbon-halide bond dissociates, forming a halogen

and a haloethyl fragment [2]. Furthermore, both the physisorbed and chemisorbed features

adsorb on Si atoms that are on neighbouring rows, rather than on the same dimer as we saw at

room temperature. We suspect that the difference in reaction is a consequence of the much

slower rate of dimer flipping at 110 K relative to the room temperature rate. We are currently

working to measure activation energies for the one-bond and two-bond dissociation reactions for

both 1,2-dihaloethanes. Future work will use DFT calculations, as implemented by VASP, to

determine physisorption and chemisorption configurations, which will then used to find the

transition state using the Nudged Elastic Band method.

[1] Harikumar, K. R.; Polanyi, J. C.; Zabet-Khosousi, A.; Czekala, P.; Lin, H.; Hofer, W.

Directed long-range molecular migration energized by surface reaction, Nature

Chemistry 3, 400-408 (2011)

[2] Huang, K., Reaction Dynamics of Alkyl Bromides at Silicon; Experiment and Theory.

Ph.D. Dissertation, University of Toronto, 2011


Growth and properties of Graphlocons (large dendritic single graphene

crystals)

Mathieu Massicotte, Victor Yu, Eric Whiteway, and Michael Hilke

Department of Physics, McGill Universisty, Montreal, Québec H3A 2T8, Canada


Graphene grown by chemical vapor deposition (CVD) on copper has attracted considerable

amount of attention due to its great potential for large scale fabrication of graphene-based

devices. However, one of the main factors limiting certain applications for CVD–graphene, is the

presence of grain boundaries of micrometer-size due to high density nucleation sites. These

domain boundaries in polycrystalline graphene reduce electronic mobilities and thermal

conductivities[1].

Hence, decreasing the number of nucleation sites and increasing the size of single grains is

important. Here, we report the growth of large dendritic-shaped and more regular shaped

graphene single-crystals (Figure 1). The crystals were grown inside a copper enclosure in order

to slow down the gas transport rate and limit the rapid surface reaction. With their sixfold

symmetry and fractal-like shape, the resulting crystals resemble snow flakes. They were found to

nucleate on copper defects and in some cases bilayers were present in the center of some

crystals. They were transferred onto a silicon oxide/silicon substrate and their lobes were

electrically contacted. The electronic properties of the devices were investigated down to sub-

Kelvin temperatures, showing mobility over 1000 cm2v

-1s

-1.

Figure 1: Graphlocon on Cu

[1] Q. Yu, L. A. Jauregui, W. Wu et al., Nature Materials 10, 443-449 (2011)


Ultra-short single-walled carbon nanotube NEMS transistors

Andrew C. McRae1, Joshua O. Island, Vahid Tayari, Serap Yiǧen and A.R. Champagne

2

Department of Physics, Concordia University, Montréal, Québec H4B 1R6, Canada

E-mails: [email protected],

[email protected]

We study electron transport in clean suspended single-wall carbon nanotube (SWCNT)

transistors hosting a single quantum dot (QD) ranging in length from 10s of nanometers down to

≈ 3 nm. Fig. 1 (a) shows a SEM image of a 70 nm breakjunction, illustrating device geometry.

To fabricate these QD devices, we align narrow gold bow-tie junctions on top of individual

SWCNTs using e-beam lithography, and suspend the devices with a buffered HF etch. We then

use a feedback-controlled electromigration to controllably open a gap in the gold junctions and

expose nm-sized sections of the SWCNTs [1]. Fig. 1 (b) shows a device with a 22 nm gap in

which the tube is visible.

We measure low temperature DC electron transport in these devices and show that they form

tunable quantum dot transistors. The Coulomb blockade characteristics show that a single QD is

housed on each nanotube. Both the stretching and bending vibronic modes are visible in

differential conductance - bias voltage - gate voltage plots of our data (Fig. 1 (c)). Despite the

bending mode‟s weak electron-vibron coupling, a high quality factor (Q) device can have a large

population of long-lived vibrons. If the current is large enough, vibron assisted electrons can then

tunnel through the QD, allowing us to directly calculate the bending frequency through DC

measurement. In our samples we find bending frequencies in the 100s of GHz range and we

calculate Q ~106. The bending frequency can controllably be tuned by tension induced by an

electrostatic or static mechanical force. These ultra-short suspended QDs are promising

candidates to develop sensitive tunable NEMS and explore the strong electron-vibron coupling

regime in SWCNTs.

Figure 1: (a) SEM image of a 70 nm nanotube connected by an electromigrated breakjunction. (b) SEM image of a

22 nm breakjunction, inset shows an enlargement of the nantotube. (c) Differential conductance data corresponding

to the device shown in (b). This data shows both stretching and bending excitations.

[1] J. Island, V. Tayari, S. Yiǧen, A. McRae and A. Champagne, App. Phys. Lett. 99,

243106 (2011)

[2] O. Usmani, Y. Blanter, and Y. Nazarov et al., Phys. Rev. B. 75, 195312 (2007)

(a)

(b) (c) (a)


A real space approach to DFT using finite-differences and LOBPCG

Vincent Michaud-Rioux, Hong Guo

University, Department of Physics, McGill University, Rutherford Building,

3600 rue University, Montréal (Québec), PQ H3A 2T8, Canada


Ab initio atomistic calculations provide a way to achieve a microscopic understanding of

observed experimental phenomena and to make quantitative predictions of the physical

properties of nanoelectronics. In practice, atomic scale systems have irregularities (e.g. surface

roughness) or defects (e.g. substitutional atoms or vacancies) that are too strong to be ignored or

treated as small perturbations. Density functional theory[1] (DFT) is a powerful method for first

principles modelling. In Kohn-Sham density functional theory[2] (KS-DFT), a system of partial

differential equations (PDE) must be solved self-consistently. In real space techniques, the KS

Hamiltonian matrix is typically much larger but also much sparser than the matrices arising in

state-of-the-art DFT codes used in the communities of condensed matter physics, quantum

chemistry, materials science and engineering. Evidence of good performance of real space

methods - by Chebyshev filtered subspace iteration (CFSI) - was reported by Zhou et al.[3] and

such real space methods are gaining popularity[4]. Some of the reasons are that real space

computational techniques are well-suited to deal with non-symmetric systems and the locality of

real space bases results in embarrassingly parallelizable algorithms.

We report the development of a real space DFT code that calculates the electronic structure of

materials. Our code is named MatRcal which stands for "Matlab-based real space calculator". All

the terms of the Kohn-Sham Hamiltonian are projected on a three-dimensional real space mesh.

Our work indicates that the performance of the locally optimal block preconditioned conjugate

gradient method (LOGPCG) introduced by Knyazev[5] generally exceeds the performance of

CFSI for solving the KS equations. We present our implementation of a LOGPCG-based real

space electronic structure calculator and use it to compute the electronic structure of many

organic and inorganic molecules. Our results are in excellent agreement with the well established

quantum chemistry commercial code Gaussian. Our method gains in computational speed and

parallelism, but the main advantage is the possibility to handle real space boundary conditions

which is critical for future applications in nanoelectronic device modeling.

[1] P. Hohenberg, W. Kohn, Phys. Rev. 136, B864–B871 (1964)

[2] W. Kohn, L. J. Sham, Phys. Rev. 140, A1133–A1138 (1965)

[3] Y. Zhou, Y. Saad, M. L. Tiago, J. R. Chelikowsky, J. Comput. Phys. 219, 172-184 (2006)

[4] L. Lin, J. Lu, L. Ying, E. Weinan, J. Comput. Phys. 231, 2140-2154 (2012)

[5] A. Knyazev, SIAM J. Sci. Comput. 23, 517 (2001)


Spin pumping/transport in magnetic metal and insulator heterostructures

Eric Montoya, Bret Heinrich, Capucine Burrowes, Bartek Kardasz, Wendell Huttema, Erol

Girt

Department of Physics, Simon Fraser University, Burnaby, British Columbia, Canada


Recently attention has been turned toward ideas where magnetization reversal and dynamics can

be achieved by spin transfer torque (STT) using spin polarized currents. STT devices are

intended for future nonvolatile Magnetic RAM and logic. Spin currents can come in two

flavours: spin polarized electric currents and pure spin currents. By using pure spin currents one

can avoid the perils of conventional electronics: circuit capacitance, heat generation, and electron

migration. In order to utilize pure spin currents in devices, a detailed understanding of spin

transport in normal metals is necessary. To study spin transport in Au, single layer

GaAs/16Fe/(20,300)Au and double layer GaAs/16Fe/(20,300)Au/12Fe/20Au heterostructures

were investigated using ferromagnetic resonance, where the numerals indicate the layer thickness

in atomic layers. By measuring the total Gilbert damping, α, in the 16Fe layer in the

aforementioned heterostructures as a function of temperature, we are able to determine the spin

mixing conductance, g

, at the Fe/Au interface and the spin flip relaxation time, τsf , in Au as a

function of temperature. We show that the spin flip relaxation time in Au is dominated by

phonon interactions [1]. Recently spin caloritronics posed the question; can one use spin

pumping to generate a spin current using magnetic insulators? Magnetic insulators, YIG in

particular, have very low magnetic damping allowing one to create a large number of low loss

magnons, a necessary requirement for an efficient spin pumping. We have shown that there is an

efficient transfer of spin momentum across the YIG/Au interface even when conduction

electrons in Au are fully reflected at the YIG/Au interface. We have shown that the spin

pumping efficiency of YIG films with an appropriate surface chemistry treatment is high and can

generate pure spin currents that are attractive for spintronics device applications [2,3].

[1] Eric Montoya, et al., J. Appl. Phys. 111, 07C512 (2012)

[2] B. Heinrich, et al., Phys. Rev. Lett. 107, 066604 (2011)

[3] C. Burrowes, et al., Appl. Phys. Lett. 100, 092403 (2012)


Vacancies in Bi2Se3: in bulk - conductance drops after 3%, in surface - robust

Vadim Nemytov, Hong Guo

Department of Physics, McGill University, Montreal, Quebec H3A 2T8, Canada


3D Strong Topological Insulator (STI) has topologically protected surface states inside the bulk

gap which can carry current or spin current.[1] Bi2Se3 has been shown to be 3D STI with a

single Dirac cone and a bulk gap of ~0.2-0.3eV.[2-4] Moreover, spectroscopic study of Bi2Se3

surface state is comparable with theoretical prediction [2-4], however transport measurements

have failed to reach the predicted value of (e2/h). [5]

In experiments, vacancies might exist in the bulk and on the surface of Bi2Se3. A preliminary

transport simulation based on finite-differences method suggests that a) vacancies at random

positions throughout the sample don't affect the conductance below about 2% vacancy

concentration (VC), beyond which the conductance starts dropping rapidly b) vacancies which

reside only on top and bottom surfaces have very little effect on conductance even for relatively

high VC. The results a) strongly suggest that the topological phase breaks in the range 3-6% VC

and that surface states disappear. The results b) may be due to robustness of the topologically

protected surface states or there may be other processes at hand; it would be interesting to

compare this to other future experiments or theoretical models.

Figure 1: Conductance in Bi2Se3 Slab due to vacancies throughout the material and only on surfaces.

[1] Liang Fu, C. L. Kane, and E. J. Mele, Phys. Rev. Lett. 98, 106803 (2007)

[2] Y. Xia, D. Qian, D. Hsieh et al., Nature Physics 5, 398-402 (2009)

[3] D. Hsieh, Y. Xia, D. Qian et al., Nature 460, 1101-1105 (2009)

[4] Y. L. Chen, J. G. Analytis, J.-H. Chu, Science 325, 178-181 (2009)

[5] Desheng Kong, Judy J. Cha, Keji Lai, ACS Nano 5, 4698-4703 (2011)


Nanomechanical resonator circuits with atomically-thin layers

D. B. Northeast, R. Knobel

Department of Physics, Queen‟s University, Kingston, Ontario, Canada K7L 3N6


Advances in nanomechanical systems [1] has allowed for the study of the motion of structures at

or near their quantum ground state for mechanical motion. Coupling these devices to resonant

electrical circuits provides not only a method of measuring with standard laboratory electronics,

but (particularly using superconducting waveguides [2]) a method for cooling towards the

ground state. Recent work [3] has recently enabled strong coupling between mechanical motion

and an LC resonator using a superconducting aluminum membrane as the mechanical element. In

this poster we propose the use of exfoliated membranes as the mechanical element, potentially

enabling a lighter resonator and stronger electromechanical coupling. Predictions based on the

use of single layers of graphene, NbSe and BSCCO are presented.

[1] J. Chan, T. P. Mayer Alegre, A. Safavi-Naeini et al., Nature 478, 7367 (2011)

[2] D. Vitali, P. Tombesi, M. J. Woolley et al., Physical Review A 76, 4 (2007)

[3] J. D. Teufel, T. Donner, D. Li et al., Nature 475, 7356 (2011)


Adsorption of meta-diiodobenzene on Cu(110): a theoretical study.

Chiara Panosetti, Werner A. Hofer

Surface Science Research Centre, University of Liverpool


We have modeled the adsorption of 1,3-diiodobenzene (meta-diiodobenzene or m-DIB) on

Cu(110) [1] by means of Density Functional Theory as implemented in VASP [2]. We have

compared the adsorption energies of 23 possible physisorption arrangements and we have

simulated STM images for the four most stable configurations using the Tersoff-Hamann

approach [3]. We discuss the stability and relative probabilities of experimentally observing the

different structures as well as the nature of the bonding (physisorption or chemisorption) with

Density of States and charge distribution arguments. We find that the adsorption induces small

distortions in the adsorbate and in some cases an adsorption-induced symmetry breakdown

occurs. Furthermore, we find evidence that the most stable arrangement is actually a bistable

system with interesting symmetry properties.

Simulated STM images of the four most stable configurations of m-DIB on Cu(110). An overlay of the

corresponding structure is shown on the right side of each panel. The images were taken at a bias voltage of -0.2 V

and plotted as isocurrent surfaces at 0.001 pA.

[1] C. Panosetti, W. A. Hofer, J. Comput. Chem, in press (2012)

[2] G. Kresse, J. Hafner, Phys. Rev. B 47, 558 (1993)

[3] J Tersoff, D. R. Hamann, Phys. Rev. Lett. 50, 1998 (1983)


Implementation of atomically defined Field Ion Microscopy tips in Scanning

Probe Microscopy

William Paul, David Oliver, Mehdi El Ouali, Till Hagedorn, Yoichi Miyahara,

and Peter Grütter

McGill University Physics Department, Montréal, Québec, H3A 2T8, Canada


The atomic scale geometry of scanning probe tip-sample junctions is usually unknown. Several

groups have investigated combinations Scanning Tunneling Microscopy (STM) with Field Ion

Microscopy (FIM) or Atom Probe (AP) methods in order to characterize material transfer during

adhesion or tip pulsing in STM experiments, but control over the exact atomic structure of the

tip-sample contact has yet to be demonstrated.

We employ FIM to atomically engineer the apex of a tungsten tip which is then used in a

combined STM/AFM. In order to preserve the atomic structure of the FIM prepared tip apex,

several considerable experimental challenges arise. We have developed techniques to preserve

the apex against corrosion by rest gases in UHV over long periods of time, and show that when

proper precautions are taken, an atomically defined apex can be approached to tunneling

proximity with a sample. We report results of recent experiments with atomically defined tips

from the tunneling to point contact regime performed at room temperature and at 150K on

Au(111) and other substrates.

These atomically defined tips are also used in the nanoindentation regime, where a precise

knowledge of tip geometry is needed to calculate contact area. From the contact area, we can

compute an upper bound for junction conductance based on the maximal conductivity calculated

for a W-Au interface, and attribute an additional conductance drop to scattering at defects and

disorder in the compressed junction.

(1) Tunneling to contact I(z) of W tip with Au(111) surface at 150K; (b) local tip changes

(red/green colour for adsorbed/evaporated atoms) due to contact with Au(111) near the center of

the FIM image.

(2) Nanoindentation regime where detailed knowledge of tip structure is used to set bounds on

junction conductivity (black lines). Supported by molecular dynamics and DFT calculations.

[1] A.-S. Lucier, H. Mortensen, Y. Sun & P. Grütter, Phys. Rev. B 72, 1-9 (2005)

[2] T. Hagedorn, M. El Ouali, W. Paul, D. Oliver, Y. Miyahara & P. Grütter,

Rev. Sci Instrum. 82, 113903 (2011)


Supramolecular Assembly of Molecules on Delta-Doped Silicon Pohl D.

1, McLean A. B.

1, Veiga R. G. A.

2, and R. H. Miwa

2

1Department of Physics, Astronomy and Engineering Physics, Queen‟s University.

2Instituto de Física, Universidade Federal de Uberlândia, Brazil.


The successful manufacture of hybrid devices, that utilize the optical response or the bio-

functional properties of organic materials [1], faces the challenge of interfacing organic material

to crystalline Si because Si still dominates the semiconductor industry. However, Si surfaces are

extremely reactive and this hinders the supramolecular assembly of organic molecules. One

approach is to passivate the surfaces with Ag or Au. Our aim is to explore the possibility of using

Si surfaces that have been doped with a single atomic layer of sub-surface B [2] as a platform for

molecular self-assembly. Our methodology is to identify molecules that will assemble into

ordered layers on the template afforded by the B-passivated surface using ab initio calculation

with density functional theory and then study the adsorption and subsequent supramolecular

assembly of these molecules with STM.

In the above figure, we show an STM image collected from a Si surface that has an incompletely

formed delta doped later. On the right, we show a unit cell of the delta-doped Si surface. The

equilibrium geometry was calculated and is in good agreement with previous calculations which

show that the surface is semiconducting and that the sub-surface S5 geometry (shown) is most

stable.

[1] J. Heath, Annu. Rev. Mater. Res. 39, 1-23 (2009)

[2] A.B. McLean, L.J. Terminello and F.J. Himpsel, Phys. Rev. B 41, 7694-700 (1990)


Thermal Conductivity of Graphene

Serap Yigen, James Porter, Vahid Tayari, Joshua O. Island, A. R. Champagne

Concordia University, Department of Physics, Montréal, Canada


We fabricated suspended graphene devices, Fig. 1(a), and measured their thermal conductivity,

, as a function of both temperature, T, and charge carrier density, n. Heat transport is a

powerful tool to obtain information about both the phononic and electronic properties of

graphene. Recent experiments on heat transport in graphene have shown a high [1], but a

detailed mapping of graphene‟s heat conductivity versus T and n is not yet available. The

measurement technique we developed is a two-point method which uses graphene as its own heat

source (Joule heating) and thermometer (resistivity). We report at temperatures ranging from

10 to 350 Kelvin, and at charge carrier densities close to the Dirac point up to about

1.51011

/cm2, in graphene crystals whose length varies from 250 nm up to one micron, Fig. 1(b).

We observed that the thermal conductivity increases by over two orders of magnitude over the

temperature range, and that it increases with the crystal‟s length [2]. can be tuned by more

than a factor of 3 with gate voltage, Fig. 1(c) [2], opening the possibility of creating room

temperature heat transistors.

1

10

100

1000

(W/m

.K)

6 7 8 9

1002 3 4

T (K)

Vgate = 0 V

l = 1 µm

l = 0.5 µm

l = 0.25 µm

1.2 μm

4

5

6

7

8

9

10

(W/m

.K)

-6 -4 -2 0 2 4 6Vgate (V)

T = 250 K

L = 0.5 µm

(a) (b) (c)

Figure 1: (a) SEM image of one of our suspended graphene device. (b) Heat conductivity, , vs T for devices

whose length varies from 250 nm to one micron. (c) vs Vgate for a 500 nm long device at T = 250K.

[1] A. A. Balandin, et al., Nano Letters 8, 902 (2008)

[2] S. Yigen, V. Tayari, J. O. Island, J. Porter, and A. R. Champagne (in preparation)


Designing the Electron and Spin Transport Properties of Molecular Nano-

Magnet Transistors

Fatemeh Rostamzadeh Renani, George Kirczenow

Simon Fraser University, Canada


A molecular nano-magnet is a single molecule that contains transition metal atoms that endow it

with a stable magnetic moment. Transistors based on single molecule nano-magnet (SMNM) are

potential candidates for spintronic devices and information storage. Theoretical studies of

molecule nano-magnet transistors (SMNMT) have been based on simple, phenomenological spin

Hamiltonians or more complete density functional theory (DFT) calculations. This talk will

report on a tight-binding model that provides a realistic description of SMNMTs comparable to

that of DFT while being capable of treating much larger molecules. Thus it allows us to propose

ways to design the properties of SMNMT. We consider in detail some examples of Mn12 based

SMNMT in which the molecules bond to gold electrodes via thiol-terminated ligands.

For neutral and negatively charged Mn12 SMNMs with acetate or benzoate ligands the model

yields the total SMNM spin, the spins of the individual Mn ions, the magnetic easy axis

orientation, the size of the magnetic anisotropy barrier and the size of the HOMO-LUMO gap

consistent with experiments.

In experiments, the orientation of the molecule‟s easy axis relative to leads is not controllable

and it has not been feasible to measure it. Our calculations reveal the possibility of determining

the easy axis orientation experimentally by means of current measurements: We find the lowest

unoccupied molecule orbitals (LUMO) of Mn12-benzoate to be on ligands, unlike the highest

occupied molecule orbitals which is located on the Mn12magnetic core. Therefore, we predict

the transport via LUMO not to be subjected to Coulomb blockade. We predict gate controlled

switching between Coulomb blockade and coherent resonant tunneling in SMNMTs based on

such SMNMs. We propose that this effect can be used to identify specific experimentally

realized SMNMTs in which the easy axis and magnetic moment are approximately parallel to the

direction of the current flow. We also predict effective spin filtering by these SMNMs to occur at

low bias whether the transport is mediated by the HOMO that is on the magnetic core of the

SMNM or by near LUMO orbitals located on the nominally non-magnetic ligands.1

Electrons can pass through the molecular nanomagnet easily only if their spins are oriented parallel to the direction

of electric current through the molecule.

[1] F. Rostamzadeh Renani, G. Kirczenow, Phys. Rev. B 84, 180408 (2011)


Fabrication of Single-electron Transistors for Charge Detection

J.P. Richard1, G. Droulers

1, M. Pioro-Ladrière

2, D. Drouin

1

1Institut interdisciplinaire d‟innovation technologique – 3IT, Université de Sherbrooke,

Sherbrooke, QC, Canada 2Département de physique, Université de Sherbrooke, Sherbrooke, QC, Canada


With the continuous downscaling of transistors, quantum effects must be considered. Many ways

are studied to exploit quantum phenomenons, especially devices that use confined electron

(quantum-dot) as the information carrier. At this scale, charge sensors are useful, for example, to

read-out the output state of a quantum system.

Ongoing studies are on their way to implement silicon quantum-dots, as well as CMOS-

compatible quantum-dot cellular automata (QCA), both based on single-electron transistor

(SET). Like quantum-dots, SET stores single charge, thus the logical result can be read-out by

charge detection devices.

Despite its transistor operation, the SET can also be used as an electrometer because of its high

charge sensitivity [1]. We present a study of the operation and the fabrication method of a

CMOS-compatible SET used as an electrometer. The fabrication is based on a nanodamascene

process [2], specifically designed for room temperature operation. We also present simulation

results of the capacitive coupling between the electrometer SET and a quantum-dot SET, as a

function of the size and the distance separating the two devices.

Figure 1 – (Left to right) Stability diagrams for zero, middle, and dominant capacitive coupling between the

electrometer SET and the quantum-dot SET. Theses diagrams were simulated using SIMON based on capacitance

values calculated with the parallel-plate capacitor model.

[1] P. Lafarge et al., Z. Phys. B. 85, 327 (1991)

[2] A. Beaumont et al., IEEE Electron Device Letters 30, 766 (2009)


Modelling of electrical percolation in size distributed carbon nanostructures

networks

Simoneau L-P., Rochefort A.

École Polytechnique de Montréal, Engineering Physics Department

Regroupement québécois sur les matériaux de pointe (RQMP)


Electronic display and photovoltaic devices typically rely on the use of transparent electrodes,

which are in their great majority rigid, brittle and expensive. An alternative that would not

possess those drawbacks is thus highly desirable and the properties of percolative networks of

carbon nanotubes (CNT) or graphene nanoribbons (GNR) place them among the most promising

candidates to reach this goal. Computational means are well suited for the description of the

underlying physics and the optimization of the characteristics of such nanostructures networks,

but previous efforts in this direction typically consider a single parameter to describe the CNT-

CNT junctions [1], which is in fact the limiting factor the propagation of charge carriers [2].

We have developed Monte Carlo (MC) algorithms to study the charge transport in bi- and

tridimensionnal networks of CNT and GNR by incorporating realistic distributions and

descriptions of CNT-CNT (or GNR-GNR) junctions. Our MC algorithms generate random

networks with many controlled parameters which can be tuned to represent experimental

networks of CNTs, GNRs or some mixture of different nanostructures. We then evaluate the total

conductance of the generated networks on the basis of individual contacts conductance, which in

turn depend on the local network properties. Our results show that the length, diameter,

orientation and chirality distributions within the percolative network of the CNT and GNR

networks have a great importance on the resulting electrical performances. By varying the

contact resistances distributions, we reproduce experimental results for CNTs networks [3].

[1] Hicks J, Behnam A, Ural A. Physical Review E 79, 012102 (2009)

[2] Nirmalraj PN, Lyons PE, De S, Coleman JN, Boland JJ. Nano Letters 9, 3890 (2009)

[3] Aguirre CM. Ph.D. Thesis (2007)


Modeling of Clocked Molecular-scale Quantum-dot Cellular Automata

Beyond the Mean-field Approximation

Marco Taucer

1, Faizal Karim

2, Konrad Walus

2, Robert A. Wolkow

1

1Dept of Physics, University of Alberta, T6G 2E1, Edmonton, AB

2Dept of Electrical and Computer Engineering, University of British Columbia, Vancouver, BC

V6T 1Z4


Quantum-dot Cellular Automata (QCA) provides a basis for classical computation without

transistors. QCA cells consist of tunnel-coupled quantum dots with mobile electrons, and

information is encoded in the positions of electrons within each cell.

In this talk, I will present computational results that treat small groups of QCA cells with a

Hamiltonian analogous to a quantum mechanical Ising-like spin chain in a transverse field. These

simulations include the effects of inter-cellular entanglement whereas many previous QCA

simulations relied upon the so-called Intercellular Hartree Approximation (ICHA), which

neglects the possibility of entanglement[1]. When energy relaxation is included in the model, we

find that intercellular entanglement changes the qualitative behaviour of the system. Specifically,

certain schemes for memory and for clocked QCA architectures, which are predicted to work on

the basis of the ICHA, do not work when entanglement is accounted for. The ICHA is a valid

approximation in the limit of very low tunneling rates, which are realized in lithographically

defined quantum-dots[2]. However, in molecular-scale QCA[3], entanglement may play a role.

The degree to which entanglement poses a problem for memory and clocking depends upon the

interaction of the system with its environment, as well as the system‟s internal dynamics.

[1] J. Timler and C. S. Lent, J. Appl. Phys. 19, 823 (2002)

[2] I. Amlani, A. O. Orlov, G. Toth et al., Science 283, 289-291 (1999)

[3] M. B. Haider, J. L. Pitters, G. A. DiLabio et al., Phys, Rev. Lett. 102, 046805 (2009)


Fabrication of nanometer-scale suspended graphene transistors

Vahid Tayari, Joshua O. Island, Serap Yiğen, James Porter, A.R.Champagne,

Department of Physics, Concordia University, Montreal, Quebec, Canada

E-mails: [email protected] , [email protected]

We present a method to fabricate suspended ultra-short graphene transistors. We define narrow

bowtie gold junctions on exfoliated graphene, and use oxygen plasma to etch away the graphene

not masked by the gold junctions. The junctions are suspended using a wet etch. We use a

feedback-control electromigration procedure to open a nm size gap in the gold bridge to expose

sections of graphene which are 100 to 300 nm wide, and as short as ≈10 nm Fig. (a). Using

electron transport, we measure conductance in these suspended graphene nanocrystals as a

function of temperature and charge carrier density Fig. (b) and observe near ballistic transport.

These ultra-short transistors offer the prospect of exploring the coupling between flexural

vibrons and charge carriers in graphene and developing ultra high frequency NEMS.

Fig. (a): Suspended gold bridge with nanometer size graphene crystal. Fig. (b):

Conductance of a ≈7nm long graphene crystal with width of ≈100nm as a function

of gate voltage which controls charge carrier density. µ=1800cm2/Vs at n=10

11/cm

2

and T=4.2K.

[1] J. O. Island, V. Tayari, et al., App. Phys. Lett 99; 243106 (2011)

[2] Hongkun Park, Andrew K. L. Lim, Jiwoong Park, et al., App. Phys. Lett 75, 301 (1999)

[3] J. Scott Bunch, Arend M. van der Zande, et al., Science 315, 5811 (2007)

Graphene

Source Drain

Gate

(a) (b)


Growth of MgO and NaCl thin insulating films on iron surfaces

Antoni Tekiel, Jessica Topple, Shawn Fostner, Yoichi Miyahara, Peter Grütter

Department of Physics, McGill University, Montreal, H3A 2T8, Canada


Thin insulating films grown on a variety of metal substrates, such as Ag(001), Mo(001) and

Fe(001), are frequently used in systems where the film serves as a tunnelling barrier directly

controlling properties of a device. For example in heterogeneous catalysis and magnetoelectronic

devices the insulating film is often made of magnesium oxide that can be grown layer by layer

allowing for control of the tunnelling barrier thickness. However, well-defined and fully

crystalline MgO thin layers are difficult to fabricate and are usually oxygen deficient. These

difficulties have been attributed to the MgO-metal lattice mismatch and the fact that in ultra-high

vacuum (UHV), regardless the preparation method, MgO is essentially grown from two separate

sources of magnesium and oxygen, which have to react on the substrate to nucleate MgO.

In this work, we characterize the morphology of MgO and NaCl thin films on Fe(001) surfaces

by UHV atomic force microscopy (NC-AFM) and low energy electron diffraction (LEED). First,

to determine the possibilities for quality improvement of MgO films grown on iron we choose

reactive deposition method, which gives full control over the gaseous species existing in the

evaporated beam. We demonstrate that it can produce much larger terraces of up to 20 nm (on an

8 monolayer thick film) compared to other methods [1,2]. Second, we investigate the

morphology and orientation of NaCl thin films grown on clean Fe(001) and oxygen passivated

Fe(001)-p(1x1)O surfaces, with the former system identified recently as an alternative tunnelling

barrier in magnetic tunnelling junctions [3]. Contrary to MgO, growth of NaCl proceeds from

molecular dimers, and thus is much better controlled due to reduced complexity of on surface

NaCl nucleation. We discuss the effects of the temperature and the oxygen presence on the

growth and demonstrate that on the oxygen passivated surface NaCl can be grown in a layer-by-

layer mode providing atomically flat films with 40-60 nm wide terraces and with much less

defects than the MgO films.

[1] M. Klaua et al., Phys. Rev. B 64 134411 (2001)

[2] M. Mizuguchia et al., Appl. Phys. Lett. 88 251901 (2006)

[3] P. Vlaic, J. Magn. Magn. Mater. 322 1438 (2010)


Short channel low field impurity mobility in graphene from first principles

Zi Wang1, Hong Guo

2, Kirk H. Bevan

1

1Department of Materials Engineering, McGill University, Montreal, QC H3A 2B2, Canada

2Department of Physics, McGill University, Montreal, QC H3A 2T8, Canada


The remarkable properties of graphene have been studied extensively since its discovery in 2004

[1,2]. Due to its very high carrier mobility, graphene has attracted considerable interest in high-

performance radio frequency (r.f.) applications [3]. Since graphene is a surface, from a

topological perspective, it will interact with any environment: substrate adatoms, imperfections,

and impurities all act as disordered scattering sites on the graphene surface, and it is believed that

defect scattering is the dominant scattering mechanism in graphene [4,5]. The present work uses

the first-principles non-equilibrium Green's function (NEGF) formalism to study the effects of

dopant disorder on carrier mobility in graphene devices. More specifically, this theory

incorporates the contributions from diffusive impurity scattering [6]. This term is typically

assumed to be negligible in transport studies of short channel devices, where current flow is

assumed to be ballistic. However, here we show that diffusive impurity scattering can play a

crucial role in the conduction properties of wide short channel device materials (such as

graphene). The results of this study could substantially impact upon the design of ultra-high

frequency graphene r.f.-devices, where short channels on the order of tens of nanometers are

needed to modulate current flow [3].

A graphene two-probe device. One possible path of an electron undergoing diffusive scattering is shown.

[1] A. K. Geim and K. S. Novoselov, Nat. Mater. 6, 183 (2007)

[2] A. K. Geim, Science 324, 1530 (2009)

[3] Y.Wu, Y.-m. Lin, A. A. Bol, et al., Nature 472, 74 (2011)

[4] K. S. Novoselov, A. K. Geim, S. V. Morozov et al., Nature 438, 197 (2005)

[5] J.-H. Chen, C. Jang, S. Xiao, et al., Nat. Nano. 3, 206 (2008)

[6] Y. Ke, K. Xia, and H. Guo, Phys. Rev. Lett. 100, 166805 (2008)

MONTREAL ATTRACTIONS

Close to McGill

Restaurants

Lola Rosa (vegetarian bistro, $12-20)

537 Milton (corner Aylmer)

Offering tasty & hearty vegetarian meals: burritos, quesadillas, lasagne, curries. Also serves

nachos and beer. One of the most tasty restaurants near McGill.

Pizza Il Focolaio ($13-20)

1223 Rue Du Square Phillips (just south of St-Catherine St)

This wood fired pizzeria always draws a large crowd - don‟t be discouraged, the service is

incredibly quick. One of Montreal‟s top pizzas.

Amelios Pizza ($10-20)

201 Rue Milton (corner Sainte-Famille)

A bring your own wine restaurant that‟s a legend in the area. Serves up well priced fresh pasta

and pizza.

Coffee

Pikolo Espresso Bar

3418 B Avenue du Parc (between Sherbrooke and Milton)

Great coffees and tasty baked snacks. A real gem in the neighbourhood.

There are also several large coffee chains nearby: Starbucks (Union/Ontario; University/de

Maisonneuve; Eaton center), Second Cup (Milton/Parc), Presse Cafe (Milton/Parc), Tim Hortons

(Parc between Sherbrooke and Milton; Sherbrooke/University)

Bars

Benelux (brewpub)

245 Rue Sherbrooke Ouest (corner Jeanne-Mance)

Benelux offers many excellent beers primarily in the Belgian and American styles.

Olde Orchard (Irish Pub)

20 Rue Prince Arthur Ouest (corner St-Laurent)

Standard Irish Pub fare.

Pullman (Wine Bar)

3424 Avenue du Parc (between Sherbrooke and Milton)

Their sommeliers will delight in finding you something really exciting to try. Grab a flight of

three wines for a tasting. Great selection of snacks and cheeses.

Liquor Stores

Beer and table wine can be bought at any grocery store or dépanneur (convenience store) until

11pm (known as “Beer O‟Clock”) in Quebec.

The SAQ provincial liquor commission has many outlets in the downtown area including one at

the Galeries Du Parc underground mall (connected to New Residence Hall).

Grocery Stores

Can be found in the Galeries Du Parc mall: Metro is a large grocery store chain. You can buy

excellent fruits and vegetables at Eden in the same mall.

Iconic Montreal Food

Smoked Meat

The Main (24 hrs)

3864 St-Laurent (between des Pins and Duluth)

Across the street from the iconic Schwartz‟, this is where real Montrealers go for their smoked

meat at any hour of the day. On the scale of lean to fatty, order a medium at the very least.

Bagels

Fairmount (24 hrs)

74 Fairmount Ouest (1.5 blocks east of Parc avenue)

Montreal‟s first bagel bakery, opening in 1919. Montreal‟s best bagel. Nearby: Dieu Du Ciel.

Rotisserie Chicken

Romados 115 Rachel Est (3 blocks east of St-Laurent)

Charcoal grilled Portuguese rotisserie chicken. Take it for a picnic in Jeanne-Mance Park, where

Rachel ends 6 blocks West of here (the City of Montreal allows the consumption of alcohol in

parks as long as one is having a picnic where picnic tables exist).

Poutine

La Banquise (24 hrs)

994 Rachel Est

This is the poutine place, if you‟re looking for it. Fries, cheese curds, and gravy of the highest

quality (100% vegetarian options available). Lots of styles to suit any poutine craving, any time

of day. Don‟t get a large unless you‟re dared to.

Beer

Dieu Du Ciel

29 Avenue Laurier Ouest (accessible from 80 du Parc bus or Laurier metro)

Montreal‟s most renowned brewpub. Always creating fantastic offerings, but can be unpleasantly

busy at night – get there early, or late (after 11pm). Fairmount bagels are just around the corner

to help soak up the excess beer for a classic Montreal night...

Places to Visit

Mount Royal

There are several convenient ways to visit the Mount Royal park from New Residence Hall. Two

popular ways to get to the Chemin Olmstead (a pedestrian & bicycle gravel road that runs

through the park) are shown in this map: http://g.co/maps/ktkep. Once you‟re in the park, you

can visit the Chalet (look-out) for a nice view over downtown Montreal. This is about 10 minutes

walk from the entrance at Peel & des Pins if you take the stairs up.

Old Montreal

Vieux-Montréal is one of the most beautiful, well-preserved vibrant “old towns” in North

America. You can get there by walking from McGill Campus through Place des Arts and

Montreal China Town. The old town starts once you cross St. Antoine Street. Walk on Notre-

Dame and St. Paul Streets from Place d'Armes to Place Jacques-Cartier. Make sure to see the

Notre-Dame Basilica, the City Hall, the Bonsecours Market and of course the Old Port of

Montréal.

Montreal Museum of Fine Arts (Musée des Beaux-Arts)

One of Canada's most famous museums, this popular institution houses a wide collection of

international contemporary and Canadian exhibits, and straddles two buildings, the 1912 original

and its 1991 across-the-street annex with underground galleries that connect the two sites.

Admission to the Museum's collection is free at all times, but the main exhibition (Beyond Pop:

Tom Wesselmann) costs, which is half price ($7.5) on Wednesday nights, from 5 to 8 p.m. for

adults. Note that the Museum's collection is closed after 5 p.m. To get there, walk west on

Sherbrooke Street. The museum is just 10 minutes walk away from McGill University campus.

http://g.co/maps/ktkep

La Ronde

Six Flags operates Canada‟s second largest amusement park on St Helen‟s Island. It is home to

10 roller coasters, and a total of 40 rides. Among them, Le Monstre is the largest wooden roller

coaster in Canada, and at 39.9 meters, currently holds the record for tallest two-track roller

coaster in the world. La Ronde can be easily reached by Metro to Jean-Drapeau station on St

Helen‟s island, where there is a bus to help get you there faster (the park is a 15-20 min walk

from the metro station). On the topic of molecular electronics, while visiting St Helen‟s Island,

go see the fabulous pentagons and hexagons of Buckminster Fuller‟s dome, built for Expo ‟67!

Olympic Stadium – Biodôme – Jardin Botanique

A Metro ride out to Viau or Pie-IX (they are very close together) will take you to these

attractions. The Olympic Stadium might not be that attractive, but the view from the top of the

tower is cool. If you‟re into observing penguins, monkeys, bats, beavers, alligators and other

animals all under one roof, check out the Biodôme where you can explore indoor replicas of four

ecosystems found in the Americas. The Jardin Botanique is just around the corner and “ranks as

one of the world‟s largest and most spectacular botanical gardens.”

Bixi

Montreal‟s self-serve bicycle rental network costs $7 for 24 hr access or $15 for 72 hr access by

credit card. Once you have access, you can borrow the bicycles as often as you wish from any

station on the network, but only for 30 minute rides: Heavy fines start accruing nearly

exponentially with time after the initial 30 minutes. For longer trips, dock the bike after short

stretches and you can take another after a 2 minute delay. If a station is full, you can get a 15

minute extension to find another one to dock the bike by inserting your credit card at the terminal

and following the directions. See the map of the Bixi network at http://montreal.bixi.com

Markets

The Marché Atwater is located at the south end of Avenue Atwater, near the Lachine Canal.

You can get to it by Bixi (check the map online for exact station locations) or by Metro to

Lionel-Groulx station. The market has many local produce vendors, restaurants, bakeries, and

butchers. A major highlight is the Fromagerie Atwater cheese shop. The Marché Jean-Talon

(accessible near Metro Jean-Talon) is much larger, with ~3x the number of vendors of glorious

local products. Lots of things to taste on site.

Tam-tams and Medieval Foam Fighters

On Sunday afternoons a large drum circle forms at the George-Étienne Cartier monument

(Mount Royal Park, on the Park Avenue side, where Rachel would intersect Park). It has

become one of the main tourist attractions of the city. People assemble in the park around the

monument to relax, tan, play frisbee, and smoke. The Medieval Foam Fighters usually battle it

out with their foam swords and spears nearby in the woods of the park. These warriors are very

kind and may offer you a sword to borrow for combat.

A

BCD

EF

A

McConnell Engineering Bldg3480 rue University, room 13

Rutherford Physics Building3600 rue University, board room

Trottier Building3630 rue University, lobby

New Residence Hall3625 avenue du Parc, lobby

Centre Mont-Royal2200 rue Mansfield

Restaurant L'Academie2100 rue Crescent

B

C

D

E

F

Rue Crescent

Rue Crescent

Mansfield St

Mansfield St

UNION

“Mon

trea

lN

orth

”

“Mon

trea

lN

orth

”

UNION

747

BU

S

PLACE DU FRÈRE-ANDRÉ

747

BU

S

PLACE DU FRÈRE-ANDRÉ

AIR

PO

RT

AIR

PO

RT

500m500m

JEANNE-MANCE

JEANNE-MANCE

Program & Abstracts - McGill PhysicsSPM/nano2012/cifar_rqmp_nano2012_progra… · Program &...

Documents

Transcript of Program & Abstracts - McGill PhysicsSPM/nano2012/cifar_rqmp_nano2012_progra… · Program &...